-
Notifications
You must be signed in to change notification settings - Fork 6
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Whitening parametrisation #60
Comments
Agreed -- this parametrisation would be very helpful. Starting with a notebook and then figuring out how it might work inside the package seems good to me. Probably we just need a way inside the |
I've also been messing around with MCMC for sparse GPs which uses the same parameterisation - so would be good to figure out a general way of handling it |
@rossviljoen Do you have an example of this somewhere? Even as a gist? |
yes, it's an example in the branch |
Am working on doing this at the minute. PR incoming shortly. |
Hahaha so we are only 3 people working on exactly the same thing 😆 |
Mine is working already 😈 |
We'll have to do double blind PR review 😆 |
In many cases, optimising an
SVGP
model is much easier when parametrising it in the "whitened" representation, i.e. forL L' = K
, we parametriseu = f(Z) = L * v
withv ~ N(0, I)
(such that under the prior,u ~ N(0, K)
is the same as before). Could maybe make use of PDMats.jl'swhiten
/unwhiten
functions. Not sure how this would work as a feature within the package, but at the very least it should be fairly straightforward to demonstrate how you would do this "by hand" in one of the notebooks.The text was updated successfully, but these errors were encountered: